FAIZ ARNI

Title: FAIZ ARNI

Research Question: How can the expressive power of logic-based languages be extended while retaining a fully declarative model-theoretic and fixpoint semantics?

Methodology: The researchers developed the LDL++ system, which includes a research project that was started at MCC in 1989. They focused on innovative and distinctive features such as new language constructs, an execution model, and an extensive application testbed. The new language constructs include choice, user-defined aggregates (UDAs), and XY-stratification. Choice constructs are monotonic and can be used freely in recursion, while UDAs are monotonic and can be identified by the compiler. XY-stratification generalizes stratification to support negation and (nonmonotonic) aggregates in recursion.

Results: The LDL++ system successfully extended the expressive power of logic-based languages while retaining a fully declarative model-theoretic and fixpoint semantics. It supported negation and aggregates in recursion, allowing efficiently even basic applications of practical importance. The system's execution model supported new language constructs, data-intensive applications, and an open architecture for extensibility to new application domains. The extensive application testbed evaluated the effectiveness of deductive database technology on data-intensive applications and new domains such as middleware and data mining.

Implications: The LDL++ system's innovative features and successful results have significant implications for the field of logic-based languages. It demonstrated that it is possible to go beyond stratification and allow negation and aggregates in recursion, opening up new possibilities for expressive power and applicability. The system's execution model and open architecture provide a framework for further research and development in the field, while its extensive application testbed serves as a valuable resource for evaluating the efficacy of deductive database technology in various applications.

Link to Article: https://arxiv.org/abs/0202001v1 Authors: arXiv ID: 0202001v1